Post on 19-Nov-2021
transcript
Low Pressure Steam Turbine - Last
Stage Blade Erosion Scarring, Notch
Removal and Refurbishments
Parsons 2019 – 10th International Charles Parsons Turbine
Conference
16-18th September 2019, Cranfield University, UK
Mehran Zanjani, Ben Morrell and Dave Carr
Uniper Technologies Limited
We are Uniper
2
Where we operate:
40+ countries around the world
4th largest generator in Europe
Employees: 12,000Our operations:
Power Generation
Commodity Trading
Energy Storage
Energy Sales
Energy Services
Power generation, Storage, Services - Europe
Power generation - International
Commodity Trading, Energy Sales
€ 1.7 bnEBITDA
100 yearsExperience
37 GW Total generation
Main activities:
Data: Uniper Annual report 2018
Gas fired plants
19.2 GW
Coal fired plants
10.5 GW
Energy storage
Gas: 8.2 bn m3
Gas pipelines and
infrastructure
Regasification
Nuclear plants
1.9 GW
Hydroelectric plants
3.6 GW
Trading Energy sales (small to large clients, electricity
and gas)
Services
Presentation Outline
Erosion scarring and notch formation on last stage blade low pressure
steam turbine
Consequences and impact of scarring and notch formation if left unattended
Last stage blade design and operation
Reasons for the formation of erosion damage
Mitigations, scarring and notch removal
Benefits gained from notch removal and blade dressing
3
Last Stage Blade Steam Turbine
4
Last Stage Blade Steam Turbine Cracking
5
A number of last
stage blades have
suffered cracking in
aerofoil trailing
edge above root
platform.
Crack Initiated from Erosion Notch
6
Causes of crack
initiation and
propagation are the
stress concentrations
associated with erosion
scarring and notches
and occasional ‘worm-
holes’.
Ultimate Failure, Aerofoil Liberation
7
When left unattended,
these cracks may
result in ultimate
failure and aerofoil
liberation, damaging
other blades, casing,
condenser as well as
risk of damage to
entire rotor train.
Erosion Damage
Erosion damage is caused by water
droplets at high speed swirling from tip to
the base of the aerofoil scarring the
trailing edge and forming macroscopic
notches and ‘worm holes’ at the base of
these scars.
8
These notches act as stress
concentrators, and in a high
stress region become
detrimental to integrity of
blade.
Last Stage Blade, Erosion Damage Causes
If unattended, the notches act as crack
initiators developing into propagating
cracks.
In areas of high mean stress (e.g. aerofoil
trailing edge immediately above the
platform) local stresses have been
calculated to increase by a factor of 2.5
resulting in crack initiation.
In combination with dynamic stresses this
could quickly develop to a blade failure.
9
Longer Blade / Greater Efficiency
Design Utopia – Achilles Heel
Drive for greater efficiency has led to longer
length last stage blades.
The increased blade size has led to higher
rotational stress
By pushing the boundaries of last stage blade
design, it has become necessary for utilities
and operators to focus on system operation to
minimise damaging factors (e.g. no / low load
operation, and spray water admission and
control).
10
High stress above the
platform on the thin
trailing edge
Higher Stresses
In some blades, the increased blade
length has resulted in higher mean
stresses, particularly above the root
platform, close to the yield strength /
ultimate tensile strength (UTS) of the
material.
Presence a notch acts as a stress
concentrator, which has been
calculated to increase stresses by
about 2.5 times, increasing stresses
above the yield / UTS.
11
Blade Stiffness and Vibration
Increased blade sizes also leads to
reduced stiffness, increasing the
possibility of blade natural
frequencies moving closer to
operating engine orders, and
increasing the dynamic stresses
around at or the positions high
mean stresses.
12
Stiffening of the blades would
require additional aerofoil
thickness however this would
increase loadings in the rotor
attachment which is another
potential area of weakness.
Main Cause of Erosion Notch Formation. Operation at Full Speed No Load (FSNL) and Cooling Water Spray
Steam Turbine Start Up:
Run-ups should be designed so that there is sufficient steam available to
synchronise when the turbine reaches full speed (UK / Europe – 3,000rpm)
and load up soon after.
Many steam turbines start from rest and can reach 3,000rpm using only one
turbine, typically the high pressure turbine on UK coal units, and intermediate
pressure on CCGT units. This results in a small amount of steam running
through the other turbines, including the low pressure turbine(s).
Once at full speed, this is referred to as FSNL and during this period, the
blades are rotating in a low vacuum condition.
High blade velocities can generate elevated temperatures due to windage
even under low pressure. Water spray systems are needed to control LP
exhaust temperatures to avoid exceeding 200°C.
13
Mitigation
Once significant cavities and notches have formed, UTG’s preferred method of
avoiding crack formation is to remove the erosion notches by dressing and
polishing affected areas of the aerofoil.
Re-designed water spray systems may help reduce the amount of ‘free’ water
and droplet sizes and reduce erosion notch formation.
14
Erosion Notch Dressing
Dressing affected areas of the aerofoil and removing the erosion notches would
avoid stress concentrations and maintain stress levels (both mean and dynamic)
close to original design.
Removal of material around the trailing edges (i.e. the thinnest part of the
aerofoil) could result in changes in the stress level in the modified area as well
as affecting the natural frequencies and vibration.
Understanding the effects of the modification is important to ensure long term
integrity of these critical rotating elements.
Finite Element (FE) modelling / analyses can be used to simulate original and
modified blade behavior and provide a good prediction of the remaining
operating life of the blade.
15
Blade Modelling and Assessment
Changes in blade vibration and stresses can be determined using finite element
model for original and modified aerofoils.
Blade geometry is easily obtained by white / blue light scanning and converting
it directly to a CAD and finite element model for analyses.
16
Finite Element Modelling and Analyses
Example of a finite element model used to determine natural frequency,
deformation as well as blade stresses.
17
Simulation of Aerofoil Metal Removal
18
Trailing edge (TE) thickness
reductions can be carried out in
stages, initially allowing a number
of iteration in stages:
1. Trailing edge thickness thinning
2. Chordwise dressing to restore
trailing edge thickness
3. Further thinning of trailing edge
thickness
Maximum levels of trailing edge
thickness thinning and chordal cut
back can be optimised by using
finite element analyses.
Typical Radial Planes for
Thinning Section
CAD modelling of actual dressing of the trailing edge (thinning).
19
CAD/FE Simulation of Modifications
Original Geometry
Thickness reduction 1
Thickness reduction 2
Stage 1 - Original Trailing Edge Thickness Cut
Back
20
Thickness reduction 1
Thickness reduction 2
Stage 1 - Modification with trailing edge
thickness reduction. Minimal chordal cut back
purely to smooth trailing edge during final
polishing.
Stage 2 - Chordwise Cutback
Stage 2 – Chordwise dressing used to restore thickness.
Blue - Original Geometry
Brown - Modified Chordwise
Cut Back Geometry
Stage 2, new profile is based on
the original TE thickness
Stage1, Reduced
Profile
Original Blade Profile
Stage 3- Chordwise Cutback and Further
Thinned
Stage 3 - The chordwise cutback profile is further dressed (thinned) to remove
the newly formed erosion notches.
Original blade profile
Stage2,
Thinned
and
chordwise
cutback
profile
Stage 3, Chordwise
cutback and thinned
profile
Stress Distribution for Stage 1 Modifications
23
Original Unchanged
Profile Trailing Edge
Original Thickness
Trailing Edge
Thickness Reduction 1
Trailing Edge
Thickness Reduction 2
\\
\\
\\
\\
\\
\\
\\
\\
\\
\\
\\
\\
\\
\
Stage 1 modification, small changes in the stress distribution around the
dressed area
Through-wall stress distributions
Stress Distribution for Changes in Stages 2 & 3Stage 2 - Chordwise Cutback to Restore the Original Thickness
Stage 3 - Chordwise Cutback and Further Thinning
24
Original Geometry stress
distribution Cutback Chordwise in Direction to
Restore TE thickness after Stage 1
Thinning
Stage 3 - Further TE
thickness reduction to
remove erosion following
Stage 2 dressing
Some increase of stress around
the chordal cutback areas of TEFurther increase of stress around
the chordal cutback and thinned
areas of TE
Comparison of Stresses between Original and
Dressed GeometriesOperating Mean and Allowable Alternating Stresses for Polished,
Good Finished and Scarred and Notched Surface
25
Normalised, Applied Centrifugal Stress (from FEA)
Original Stage 1 Stage 2 Stage 3
Good Finish 1.000 1.029 1.027 1.021
Erosion Notches 1.111 1.114 1.114 1.113
Normalised, Allowable Alternating Stress
Original Stage 1 Stage 2 Stage 3
Good Finish, Polished Surface 1.000 0.886 0.894 0.918
Erosion Notches 0.310 0.302 0.302 0.306
Conclusions
Erosion notches due to water droplet erosion is detrimental on highly stressed
low pressure turbine blades.
Alternating dynamic stresses as well the elastic mean stresses at the base of
the notches have been calculated to increase by 2.5 times.
Dressing of the training edge will remove the stress raisers in already near yield
strength area, increasing the remaining life of the blade.
Dressing the blades on the suction surface needs to be undertaken in a
controlled manner and, ideally, to be supported and verified by analysis.
Analysis is an efficient way of determining the acceptable limits of thickness and
chord reduction.
26
Thank you!
If you need any further information, please contact:
Mehran Zanjani Ben Morrell
Structural & Flow Analysis Steam Turbines and Auxiliaries
T: +44 (0)7972-217713 T: +44 (0)7976-466454
E: mehran.zanjani@uniper.energy E: ben.morrell@uniper.energy
Uniper Technologies Ltd
Technology Centre
Ratcliffe-on-Soar
Nottingham NG11 0EE
UK
www.uniper.energy
Uniper disclaimer:
This presentation may contain forward-looking statements based on current assumptions and forecasts made by Uniper SE management and other information
currently available to Uniper. Various known and unknown risks, uncertainties and other factors could lead to material differences between the actual future results,
financial situation, development or performance of the company and the estimates given here. Uniper SE does not intend, and does not assume any liability
whatsoever, to update these forward-looking statements or to conform them to future events or developments.